OK i need youre help eople why doesnt this thing work on this setting

[residentevil2]

Ok forget the other threads about this cpu not working on 2.2ghz i found out why.

Ok first i went back to 2ghz stock everything. When i go to 1.72 vcore or higher even on stock cpu speed when i try to do prime95 or wcpuid or even play a game like UT2K3 the computer failes giving me a blue screen.


I found out its a voltage problem. The cpu would give me 2.4ghz or 2.3 ghz if only the thing would function right on .72 vcore or higher.

Why does this cpu not work right in 1.72vcore.

Is the cpu just bad or could it be something else.
My 2000+ worked fine in 1.90 vcore only when i go to 2v it does the same thing that the 2400+ does in 1.72v.

What should i do this is making me made i know the cpu has more potential than what i'm getting now the only reason why im stuck at 2.15ghz is cuss when i go higher i need more voltage but when i give more voltage it gives me errors.

So how am i ganna fix this.

[daS_mAgg0t]

The voltages wont give you errors, its the extra speed your bringing with it thats the issue, trying running at the same speed but the higher voltage. See if your errors still occur.

[seoultrain]

Originally posted by residentevil2
Ok first i went back to 2ghz stock everything. When i go to 1.72 vcore or higher even on stock cpu speed when i try to do prime95 or wcpuid or even play a game like UT2K3 the computer failes giving me a blue screen.


I found out its a voltage problem. The cpu would give me 2.4ghz or 2.3 ghz if only the thing would function right on .72 vcore or higher.

Why does this cpu not work right in 1.72vcore.
.

That's exactly what he just did...

and btw, speed by itself rarely causes problems, unless you're so fast that you're outrunning your hard drives

I wish i knew enough to help you out, cuz it's definitely a problem with your chip not being able to handle higher voltages... good luck

[daS_mAgg0t]

"cuz it's definitely a problem with your chip not being able to handle higher voltages"

/ME Smirks in an arogant manner and restrains himself from commenting

[daS_mAgg0t]

I think the best thing to do atm, imo, which im affraid is a pain in the ass, is to take all ur devices bar AGP card out, and disconnect all unessential peripherals. Use only 1 memory chip..../me glances at your MB...and put the single memory chip in slot 2

Start from there.

[rippedsocks]

stole the words from my mouth maggot

[residentevil2]

lol

[Demont]

Originally posted by seoultrain


That's exactly what he just did...

and btw, speed by itself rarely causes problems, unless you're so fast that you're outrunning your hard drives

I wish i knew enough to help you out, cuz it's definitely a problem with your chip not being able to handle higher voltages... good luck

the harddrives thing, was that a joke?? :confused:

[seoultrain]

nah, not a joke... i actually heard of it happening.. don't really remember any details, but it was the only time i could think of when a computer was TOO fast

[Dragon_Engineer2]

Originally posted by daS_mAgg0t
The voltages wont give you errors, its the extra speed your bringing with it thats the issue, trying running at the same speed but the higher voltage. See if your errors still occur.

Correct me if I'm wrong seoultrain but what your trying to say is that his fsb is to high for his hard drive and causing the system to become unstable?If thats what your saying then you would be correct about high fsbs causing the hard drive to become unstable.However high fsbs are not the issue.

[seoultrain]

i'm not trying to say anything... the whole hard drive thing is unrelated to residentevil's problem, i'm just saying in general, "speed by itself rarely causes problems"... i'm sorry if i threw people off

[daS_mAgg0t]

Well yes i'd agree as he says he having the same issues at stock speeds, btw..was a quote to my post in reference to that...or are you a fan of the mAgg0t :beer:

[water_cooler 20]

maybe your mb is just not made for a xp 2400

[Dragon_Engineer2]

Lol that was supposed to be a quote from seoultrain.

[daS_mAgg0t]

lol :p

[daS_mAgg0t]

"maybe your mb is just not made for a xp 2400"

You'll find that most MB's won't have a problem running the CPU's coming out atm. There the same pin arrangment, therefore the same slot..i.e. slot A

MB's normally just need a software based 'flash' upgrade, which updates your motherboard to support the new multiplier, and to 'teach' it to automatically recognise a chip. But issues are normally software based..and fixed very swifty. For instance, my board, as i said in another post, is nealry over a year old now. And they just released a new bios to allow me to use a 2400+/2600+ processor :D

[Demont]

Originally posted by seoultrain
i'm not trying to say anything... the whole hard drive thing is unrelated to residentevil's problem, i'm just saying in general, "speed by itself rarely causes problems"... i'm sorry if i threw people off

if you are not joking i would really like to see a link to this info or something. ive never heard of this phenomenon of the computer going too fast for the harddrive. to be honest i think you are mistaken, but who knows. :)

[daS_mAgg0t]

I think he is referring to boards without a 1/5 divider, such as mine, or boards without a 1/6 at the higher end of the spectrum, in relation to extremely high FSB settings causing the IDE to fall out of spec, which I agree, can cause HD errors/restarts and ever corrupt your win install

He speaks the truth, its real, he just speaks it in an incomprehendable format :p

[Edit] "Athlon XP 1600+@2100+ - [AROIA-Y][138x12.5]......."
"ive never heard of this phenomenon of the computer going too fast for the harddrive"

At those speeds, no doubt you never will !!!

/me creases up in a fit of laughter :p

[Demont]

Originally posted by daS_mAgg0t
At those speeds, no doubt you never will !!!


first of all, that was a double negative, secondly, if you meant to say "ever" instead of "never" then that is an insult, and something that is 1. against the rules here, and 2. frowned upon by all members. ive seen you in a couple threads here and there recently acting like this, hopefully you will soon learn that inflamatory behavior is not accepted here. before there are any conseqences.

secondly, since i am an insecure person, i might as well tell you that i bought this board the better part of a year ago and it turned out to be a dud once i tried to overclock it later. ( i was a bit of a newb at the time.) so thats about all the fsb i can get. leaving me with only being able to overclock with the multiplier primarily. until such time that i have enough money to buy a board and a board comes out that interests me (come on nforce2). i will just stick with this board, because i see no real reason to waste money on some decent board that i will just want to replace if nforce2 turns out to be awesome. happy?

[residentevil2]

guess were getting outa subject here

<------- soppose to help me :D

ok im on stock speeds 1.72vcore wont work.


i cleaned the whole cpu cleaned the pits out all of them with a little tiny thumb tab pin. then i cleaned the cpu with a tooth brush and a little water. then used and erraser to clean the burn marks from the laser cutting the bridges.

Turnd the computer on put on 1.80vcore went into windows xp. WOW it was a little slow tho but 5 min later it locked up and i couldnt get in cept by using 1.70vcore or less. so somethings wrong with the cpu i just dont know what.

How could cleaning the cpu like that alow me to get in win xp for while

[Demont]

Originally posted by residentevil2
guess were getting outa subject here

<------- soppose to help me :D

ok im on stock speeds 1.72vcore wont work.


i cleaned the whole cpu cleaned the pits out all of them with a little tiny thumb tab pin. then i cleaned the cpu with a tooth brush and a little water. then used and erraser to clean the burn marks from the laser cutting the bridges.

Turnd the computer on put on 1.80vcore went into windows xp. WOW it was a little slow tho but 5 min later it locked up and i couldnt get in cept by using 1.70vcore or less. so somethings wrong with the cpu i just dont know what.

How could cleaning the cpu like that alow me to get in win xp for while

too true, sorry, i will remove my post once the flame goes away.

for your problem though, how do you *KNOW* its the cpu? have you tested a different cpu in there? that might be a good idea before you work at getting a replacement for that one.

[residentevil2]

yesi tried my 1.2 duron and the xp 2000+

[Demont]

Originally posted by residentevil2
yesi tried my 1.2 duron and the xp 2000+

sorry i didnt catch this in your first post:
My 2000+ worked fine in 1.90 vcore only when i go to 2v it does the same thing that the 2400+ does in 1.72v.

so the problem(s) are not entirely localized on the new cpu. perhaps something else is the problem? (motherboard?)

oh... um.... btw, the default max voltage for that board is 1.85 correct? so you either disabled the voltage limiter option or you did a volt mod..... maybe that is the problem?

[seoultrain]

are you completely locked again? if not, then give it another cleaning to make sure your unlocking isn't causing the problems

[rippedsocks]

play with the fsb as much as u can on default settings (clean that cpu!)... the boys over at hard ocp said they played with the fsb and managed to get the same overclocks as with lower multiplyers. according to them with there slk-800 and vantec tornado they managed to get it to 2.5 without unlocking the chip thats 15x166 respectively try that out

[residentevil2]

my board givesm e a overvolt jumper so dont have to mod it to get 2 v

[Demont]

thats what i meant by disabling the voltage limiter. perhaps this thread will be relevant to this problem or at least something.
http://forum.oc-forums.com/vb/showthread.php?s=&threadid=127870

[residentevil2]

disabling the voltag limiter u mean the jumper that stops wild voltage from going through the computer or do you mean going into the bios and turning of the voltage moniter ?

[papounet]

dear

I went to www.amd.com and looked at the docs
XP Model 8 specifications (http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/25175.pdf)


Table 2. Electrical and Thermal Specifications for Processors with a CPUID of 681
Frequency in MHz
(Model Number)
VCC_CORE
(Core
Voltage)
ICC (Processor Current)
Thermal Power5 Maximum Die
Temperature Working State C0 Stop Grant S11, 2, 3, 4
Maximum Typical Maximum Typical Maximum Typical
1667 (2000+) 1.60 V 38.3 A 34.8 A 8.1 A 4.9 A 61.3 W 55.7 W 90°C
1800 (2200+) 1.60 V 39.3 A 35.6 A 8.1 A 4.9 A 62.8 W 57.0 W
85°C
2000 (2400+) 1.60 V 40.8 A 37.1 A 8.1 A 4.9 A 65.3 W 59.3 W
1.65 V 41.4 A 37.6 A 8.9 A 5.4 A 68.3 W 62.0 W
2133 (2600+) 1.65 V 41.4 A 37.6 A 8.9 A 5.4 A 68.3 W 62.0 W

this mean that the higher end XP are rated for running at voltage 1.65V max. This value is not arbitrary and fixed by AMD to limit our ability to overclock. It is the average value for which the CPU is garanteed to work. It is slightly more that what some CPU may need (thus you can undervolt some), but not by much as excessive voltage cause heat and electro-migration.
this max voltage is linked to the design (.13 micron ??), material (silicon with copper interconnect) and processes used

Please bear in mind that all CPU of the same batch do not have the same capabilities. Something that can work for someone, can fail for you

It seems that you have a so-so CPU, which does not take more .

There is not much thing you can do, Sorry, except eprhaps attempt burn-in : continous operation for 1 week at you max voltage with a CPU excerising program (seti would be OK,)

[residentevil2]

Wow now thats the best explanation i have heard all day thanks alot :D.

Youre right i think i do have a so-so cpu that only works on what it was ment to and that makes me mad.

I guess ill just rma it make a up a story get the other cpu and hope that that one wont be like this cpu .

[papounet]

I read one of the other post (yours ?) regarding Amperage.

Those babies really like power.
If you look into the table, the max amp is the second figure,
then consider how much more amp is need for each increase in speed.
My own computation sin my head tell me +/- 1 amp per 100 Mhz (for those CPU at that speed),
this would mean to move you 2400+ (real MHZ 2000) to real 2200 MAY require max amp41.4 + 2 = 43.4.

But this power is not from the 5V, it is converted by the power regulator from the 12V
check page 13 of http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/26003.pdf

Model 6 AMD Athlon XP1800+ (1.533GHZ, 1.75V) draws 6.87 A of 12V for a total of 82.47 W

2. To calculate the processor current at 12V VRM source:
I= (Processor Core Voltage x Processor Core Current ) x12 x1.25


this mean that your model 8 AMD XP2400+ (2000MHZ,1.65V) draws 7.11 A of 12V for a total of 85 W

worst case : both Amp and voltage increase needed, i'll extrapolate that
your model 8 AMD XP2400+ (2200MHZ,1.72V, 43.4 A) draws 7.77 A of 12V for a total of 100 W

before RMA the poor CPU, i'll check the numers on the Powersupply.
if you push

[residentevil2]

so what are you saying ? exsacly that my pus isnt giving enough voltage for the cpu?

My +12 is only 16amp

[papounet]

It seems that there are 2 different XP 2400+ :
one at 1.65V, one at 1.6V.

my guess: the one with 1.6V is the latest stepping

could you confirm the default voltage of your CPU ?

+ 1 additionnal question :
when you report setting 1.72V, is it the value set in the bios or the one reported by the HW monitor / MBM

[residentevil2]

the one in the bios hw reports 1.75 but bios is set to 1.72


hrmm myn does 1.65v stock voltage but my stepping is newer than the reveiwing cpu's wich have

AIUAB

Myn is

AIUCB

[residentevil2]

ahh so isee this could be due to my psu right ?

Youre saying that on 2200mhz 1.72v is 43.4amp (+5vline) and uses 7.77 amps on my 12v line.

when my psu only puts of 40amp +5 and 16 amp 12v.


This means that i need a pus that either puts of more amps on the +5 line or the 12v line or do i need both.

So my problem sounds psu related ??? whats 2000 mhz and 1.72v why doesnt that work .

Also i tried 14x133=1862mhz and 1.72 v still didnt work

[papounet]

I think that my earlier indication still sticks :
you got a so-so CPU of one of the earliest batch, which historically never perform as well as later one.

BUT, As I was intrigued by the notion entertained elsewhere about 5V rail, I decided to check the data. thus I went to AMD and did the computation.

please check the total budget equation in the AMD doc
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/26003.pdf

You'll see the max current at each voltage level for a typical system
M/B .3
system fan ..25
processor fan .25
IDE HD 2
CD R/W .8 total for device 3.6 ,
but apply correction factor of 80% as per AMD recommendation, total current 2.88 A on the 12V
plus the 7.7 A for the CPU at 2200Mhz 1.72V
this mean the total load is 10.65 A on the 12V line

Which your PS should be able to supply.

BUT you may have lot more fan and HD,
and your previous attempts at overvolting your other cpu were thwarted for similar reason.

Hmm, I'll redo the computation for your previous CPU
XP 2000+ (1667MHZ, 1.6V, 38.3A)
already 38.3A at 1.9V would draw 7.58 A
at 2V would draw 7.97 A

Please tell me which speeds you attained with both your old CPU and you old, and which other peripheral uyou have in your system,
because we can't eliminate YET your PS from capping your attempts.




I

[residentevil2]

ok this is what i have hooked up now.

My 9700 pro radeon

My maxtor ata 133 7200 40 gb drive

Cdrom, DvD, Floppy

2x256pc 2700

2 small fans not sure how big but the size of a north bridge on the mobo.

The psu has 2 fans plugged up

Asound fast ethernet card and thats it.


Now my xp 2000+ i got to 2ghz using 2.14v

My 2400+ i can only get 2150mhz stable but using windows 98 since it actually boots up but has errors i can get 2350 mhz on 2v.
But like it is it isnt stable on 1.70v or more

[papounet]

as a side note, I remember from reading on the net (probably on www.overclockers.com) that someone solved some of his problem with power supply and amperage by getting rid of the plug to connect the CPU to the M/B, because contact isn't always perfect and those plugs just not good enough for large amperage. (this is why Intel changed the layout of the plugs for P4 system

[residentevil2]

danm you posted right after me well read up top what i wrote

[residentevil2]

LOOK here go down the page a bit and youll see a pic of the guys cpu.


This is exsacly what i have the same numbers same stepping same 0239 same everything.

and his works fine on 1.70 and higher even 2 v.


So it might be my psu or it might be the cpu but ithink its the cpu.

Cuss i wen to 13x133=1730 mhz and tried 1.75vcore it wouldnt work.

I think the cpu is flakey or something what do you think ?


http://forum.oc-forums.com/vb/showthread.php?s=&threadid=127604

[OC Detective]

Is your water cooling setup (in your sig) powered by your PSU or separate? I assume your 3.3V and 5V readings are stable at all times?

[residentevil2]

yes they are and no its not my pump is a ehiem 1048 wich pluggs into the wall.


So you saw the pic right that cpu all those numbers are the exsact same as myn AIUCB stepping everything from 0239 all the same .


So my cpu isnt a bad stepping i just prob got a flakey cpu and need to rma it but first im waiting for the other guy to repost

[OC Detective]

Clutching at straws now but is there any way that your mobo might be adversely affected by having 2.14V put through the CPU (your old XP2000+)?
As I have said before stepping is only an indication of whether it will overclock well - there are no guarantees - I dont think you should give up just now on this one!

[residentevil2]

why not give up iv tried everything for 2 days straight and im getting tiredo f it .


Btw i put my xp 2000+ back 2 v ran fine 1.85v ran ine its this damn cpu

[papounet]

I may be walking on a limb farr ouuut there, but let's see

Originally posted by residentevil2
ok this is what i have hooked up now.

>9700 pro radeon :
doesn't use 12V

>maxtor ata 133 7200 40 gb drive
WWW. mwaxtor.com gives max 434 mA for 12V for seek operation for Diamonf max plus 8 model

>Cdrom, DvD,
Amd says .8 A for each

>Floppy
>2x256pc 2700

>2 small fans not sure how big but the size of a north bridge on the mobo.
typical fans draw .24 A

>The psu has 2 fans plugged up
should not count

after factoring in 7.5 we are still below 12A

>Now my xp 2000+ i got to 2ghz using 2.14v
XP200+ is 1667Mhz, 1.6V, 38.3 A
I guess XP2000+ at 2000Mhz at 2.14V, 41.3 A (because of my assumption of 1A per 100Mhz) would draw
2.14 * 41.3 / 12 * 1.25 = 9.2 A

So if you manage to run stable your 2000+ at 2000MHZ with 2.14V, it is not the max amperage on the 12V line of your PS which is preventing you from pushing your 2400+ further.
So no need to go and buy another PS, it won't help.

You may nonetheless plug you CPU into another MB with another PS to double check (and be ready to cry)

> My 2400+ i can only get 2150mhz stable but using windows 98 since it actually boots up but has errors i can get 2350 mhz on 2v.
But like it is it isnt stable on 1.70v or more

Again, pushing more voltage than planned into a CPU helps overclocking because it makes the signal stronger, so easier to detect over the noise. H

Your own experience with one of those early CPU just tells you that however strong the voltage you throw, there is enough electrical noise because of the speed to cause random error, which crash your system.

Please remember the Pentium III 1.3Ghz fiasco, (Yes, that was one of the last interesting piece from THG). A CPU which is pushed beyond the process capability will fail if excercised in a certain fashion.

I hoped to have help you understand the complex issues of which I have only a partial understanding myself.

[papounet]

In case the serie of posts were not clear,

my original thought was :
early batch, so-so chip => too bad for the guy , try to RMA it

then I read a post about the 5V rail, which is incorrect
so I looked up the references
found data on the 12V load for the exact CPU, and in the end, almost eliminated the P/S from being the limiting factor.

my conclusion is still
early batch, so-so chip (unable to accept +10% Voltage) => too bad for you, try to RMA it